BatchCrypt - Enhanced "Batch Zero" Version

This repository contains an enhanced implementation of BatchCrypt based on the ATC'20 paper:
"BatchCrypt: Efficient Homomorphic Encryption for Cross-Silo Federated Learning"
Original authors: ATC'20 Paper

Enhanced by Kunha Kim, Hankuk University of Foreign Studies.

🔧 Features

• Implements a novel "Batch Zero" optimization that skips encryption of all-zero gradient blocks.
• Demonstrates significant reductions in training time with minimal impact on accuracy.
• Includes quantization, stochastic rounding, and homomorphic encryption based on the Paillier cryptosystem.

🧪 Recommended Setup

• Python 3.7+
• TensorFlow 2.x